safe-rlhf
Robustness and stress testing
Safe RLHF: Constrained Value Alignment via Safe Reinforcement Learning from Human Feedback
Contribute
Spot an error, or know a record we should add? Members can propose additions and submit corrections.
Regulation TrackerLaws, orders and treaties read for their security duties and clocksStandards & FrameworksAttack taxonomies, control catalogues and frontier frameworksAI Incident DatabaseThe security incident record, with mechanism and lessonResearch Paper LibraryThe papers that define the field, summarisedOSS DirectoryRed teaming, guardrail, supply chain and evaluation toolsCase StudiesProgrammes built, with measured outcomesAI Vulnerability TrackerCVEs and advisories for the components AI systems are built from